Conservation Evidence strives to be as useful to conservationists as possible. Please take our survey to help the team improve our resource.

Providing evidence to improve practice

Individual study: The effect on vegetation of removing non-native European rabbits Oryctolagus cuniculus from the Todos Santos islands, Baja California Sur, Mexico

Published source details

Donlan C.J, Croll D.A. & Tershy B.R. (2003) Islands, exotic herbivores, and invasive plants: Their roles in coastal California restoration. Restoration Ecology, 11, 524-530

Summary

The presence of exotic herbivores and invasive plants are generally considered to pose threats to ecosystems of the Pacific islands off southern California, U.S.A. and Baja California, Mexico. In this study, introduced herbivores were removed as a large-scale experimental manipulation to examine the importance of grazing processes on a large-scale restoration effort. A paired approach was used on the Todos Santos Islands, Mexico by removing herbivores from one island, while they temporarily remained on a similar adjacent island. Smaller scale herbivore exclosures on the control island were also used to assess the effect of non-native grazers on the vegetation.

Study site: The herbivore removal experiment was conducted on The Todos Santos Islands, two islands approximately 90 km south of the US–Mexico border 10 km off the coast of Baja California Sur, Mexico. Most precipitation (approximately 90%) falls between November and April, with an average annual rainfall of 255 mm. In this study, herbivores were removed Todos Santos South (TSS 1 km²) and Todos Santos North (TSN 0.3 km²) was used as the control. The two islands are separated by approximately 200 m and are similar both in vegetation and fauna. They are floristically diverse with 142 vascular plant taxa.

At the beginning of the study feral cats Felis catus and European rabbits Oryctolagus cuniculus were present on TSS and cats, rabbits and donkeys Equus asinus on TSN. Nonnative plants have recently become abundant, with 34 species recorded in 1994 compared to just nine in 1950.

Herbivore removal: Non-native herbivore removal from Todos Santos was a part of a regional island conservation program. Rabbits and cats were removed from both islands but removal on TSN was delayed to facilitate this experiment. Removal efforts on TSS began November 1997 with approximately 30 cats and 40 rabbits removed (pre-removal rabbit density estimate of 40/km²). Most rabbits (>95%) were removed between November 1997 and January 1998, with the last eradicated in spring 1998. Cats, rabbits and four donkeys were present on TSN throughout the study. Approximately 30 rabbits were removed from TSN at the completion of this study (pre-removal density estimate of 100/km²). The comparison of rabbit density estimates are not ideal because they are from different time periods, however, when combined with comparative observations TSN appeared to provide a valid control. Given the small numbers of donkeys (one to four) present on TSN both historically and just prior to the study, rabbits were considered the main vertebrate herbivore on both islands.

Plant cover and diversity: Plant community changes were measured along randomly positioned 100 m transects on both islands over a 17-month period and within herbivore exclosures and control plots on TSN. Vegetation transects on TSS (n = 30) were permanent but due to access restrictions on TSN, new transects (n = 20) were sampled during each sampling visit. On TSS transects were sampled once before (November 1997) and three times after (April 1998, December 1998 and April 1999) the herbivore removal. On TSN transects were sampled within 2 weeks of those on TSS, except for the first period when TSN was not surveyed.

Using the line-intercept method percent cover and species richness for each transect during each sampling period was estimated. Only plant species found on both islands were included for between-island species richness comparisons.

Herbivore exclosures: Three randomly located, 27 × 27 m exclosures were constructed on TSN in May 1998. Exclosures were made of metal fence posts, chicken wire and barbed wire, the mesh fence dug in 0.5 m deep (to prevent rabbit entry) and 1.25 m high. Control plots (also 27 × 27 m) were located 25 m from the exclosures. There was no evidence of rabbits or donkeys breaching the exclosures. The exclosures and control plots were sampled along five 25 m transects in each. Transects were at least 2 m away from exclosure edges to minimize edge effects. Percent cover (line-intercept method) estimates were made in July 1998, December 1998 and July 1999.

Plant diversity and cover: During winter 1998 heavy rains (over twice the average) fell between September 1997 and March 1998, coinciding with the start of an El Niño. After March 1998 rainfall declined to normal levels with 232 mm of rain from April 1998 to April 1999.

Patterns of species richness and cover were similar on the islands, peaking in spring and reflecting the amount of precipitation. Species richness was comparable throughout the study, except in April 1998 where TSS diversity was slightly higher. Vegetation cover was highest on both islands during April 1998, corresponding with high rainfall. During April 1999 cover on TSS (herbivores removed) was infact less (around 65%) than TSN (>100%).

Native plant cover showed little seasonal change but exotic cover changed dramatically. Native cover on TSS was consistently higher than TSN with perennials dominating. Except for the annual wild cucumber Marah macrocarpus (5–8% cover), native annuals were rare (< 2% cover). With the exception of wild cucumber on both islands and two annuals on TSS, changes of native annual and perennial cover were not significant over time on both islands (six perennials and six annuals on each island). In contrast, cover changes of all exotic annuals (with the exception of iceplant Mesembryanthemum crystallinum) were significant over time on both islands (five species). During April 1998 exotic annual cover on both islands was more than 100%, and but in December declined to < 1%. During April 1999 exotic cover on TSS was low (28%) compared with TSN (80%).

Negative correlations between native and exotic cover were apparent, primarily during spring sampling periods. On TSN in both spring 1999 and 2000 when there was high exotic cover, native and exotic cover was negatively correlated. On TSS this effect was only apparent in spring 1999.

Herbivore exclosures: On TSN, no differences in cover between the exclosures and control plots were apparent. In one exclosure and one control plot, cheeseweed Malva parviflora was the only species present with high cover in spring (>100%), dying back during December 1999 (<6%). Other exclosure and control plots consisted of both native and exotic species where vegetative cover changes were similar.

Conclusions: On the Todos Santos islands no herbivore effect on the plant communities was apparent. Plant community dynamics appeared to be dominated by El Niño related rainfall and non-native annual plant species.


Note: If using or referring to this published study, please read and quote the original paper, this can be viewed at: http://www.blackwell-synergy.com/journal.asp?ref=1061-2971